Intersubunit rotation in active F-ATPase

The b and delta subunits of the Escherichia coli ATP synthase interact via residues in their C-terminal regions

An affinity resin for the F1 sector of the Escherichia coli ATP synthase was prepared by coupling the b subunit to a solid support through a unique cysteine residue in the N-terminal leader. b24-156, a form of b lacking the N-terminal transmembrane domain, was able to compete with the affinity resin for binding of F1. Truncated forms of b24-156, in which one or four residues from the C terminus were removed, competed poorly for F1 binding, suggesting that these residues play an important role in b-F1 interactions. Sedimentation velocity analytical ultracentrifugation revealed that removal of these C-terminal residues from b24-156 resulted in a disruption of its association with the purified delta subunit of the enzyme. To determine whether these residues interact directly with delta, cysteine residues were introduced at various C-terminal positions of b and modified with the heterobifunctional cross-linker benzophenone-4-maleimide. Cross-links between b and delta were obtained when the reagent was incorporated at positions 155 and 158 (two residues beyond the normal C terminus) in both the reconstituted b24-156-F1 complex and the membrane-bound F1F0 complex. CNBr digestion followed by peptide sequencing showed the site of cross-linking within the 177-residue delta subunit to be C-terminal to residue 148, possibly at Met-158. These results indicate that the b and delta subunits interact via their C-terminal regions and that this interaction is instrumental in the binding of the F1 sector to the b subunit of F0.

ATP synthase--past and future

This paper gives an overview of a lecture scheduled for the opening of the 10th European Bioenergetics Congress. In this lecture I plan to first reflect on the accomplishments of some of the individuals who were involved in research on the ATP synthase during the past 50 years. Then I will give a brief view of the present information about rotational catalysis by the ATP synthase. This will be followed by a discussion of some results from my laboratory that call for additional experimentation. Finally I will direct attention to other questions about the ATP synthase that should be addressed in future studies.

The motor of the ATP synthase

A model is presented in which ion translocation through the F0 part of the ATP synthase drives the rotation of the ring of c subunits (rotor) versus the a subunit (stator). The coupling ion binding sites on the rotor are accessible from the cytoplasm of a bacterial cell except for the c subunit at the interface to the stator. Here, the binding site is accessible from the periplasm through a channel formed by subunit a. In the ATP synthesis mode, a coupling ion is anticipated to pass through the stator channel into the binding site of the adjacent rotor subunit, following the electrical potential. Occupation of this site triggers, probably by electrostatic forces, the rotation of the ring. This makes the binding site accessible to the cytoplasm, where the coupling ion dissociates. Simultaneously, this rotation moves again an empty rotor subunit into the contact site with the stator, where its binding site becomes loaded and rotation continues.

Electron microscopic evidence of two stalks linking the F1 and F0 parts of the Escherichia coli ATP synthase

The structure of monodisperse ATP synthase from Escherichia coli (ECF1F0) has been examined by electron microscopy after negative staining of specimens. The F1 part is seen to be connected by two stalks. One is more centrally located and includes the gamma and epsilon subunits. The second stalk, observed here in ECF1F0, is arranged peripherally. It probably contains the delta and b subunits which, in addition to gamma and epsilon, are required for binding of the F1 and F0 parts of the complex. Other novel features of the F1F0 complex can be discerned. There is a cap at the top of the F1 part at which the second stalk may bind. This likely includes N-terminal stretches of the three copies of the alpha subunit and a part of the delta subunit. The F0 part is clearly asymmetric. The presence of two stalks in the complex has important functional implications. There is good evidence that the more central stalk of gamma and epsilon subunits is a mobile domain that rotates to link the three catalytic sites on beta subunits in turn, with the proton channel of the F0 part. The second stalk of delta and b subunits is then the stator which makes this rotation possible.

Interacting helical faces of subunits a and c in the F1Fo ATP synthase of Escherichia coli defined by disulfide cross-linking

Subunits a and c of Fo are thought to cooperatively catalyze proton translocation during ATP synthesis by the Escherichia coli F1Fo ATP synthase. Optimizing mutations in subunit a at residues A217, I221, and L224 improves the partial function of the cA24D/cD61G double mutant and, on this basis, these three residues were proposed to lie on one face of a transmembrane helix of subunit a, which then interacted with the transmembrane helix of subunit c anchoring the essential aspartyl group. To test this model, in the present work Cys residues were introduced into the second transmembrane helix of subunit c and the predicted fourth transmembrane helix of subunit a. After treating the membrane vesicles of these mutants with Cu(1, 10-phenanthroline)2SO4 at 0 degrees, 10 degrees, or 20 degreesC, strong a-c dimer formation was observed at all three temperatures in membranes of 7 of the 65 double mutants constructed, i.e., in the aS207C/cI55C, aN214C/cA62C, aN214C/cM65C, aI221C/cG69C, aI223C/cL72C, aL224C/cY73C, and aI225C/cY73C double mutant proteins. The pattern of cross-linking aligns the helices in a parallel fashion over a span of 19 residues with the aN214C residue lying close to the cA62C and cM65C residues in the middle of the membrane. Lesser a-c dimer formation was observed in nine other double mutants after treatment at 20 degreesC in a pattern generally supporting that indicated by the seven landmark residues cited above. Cross-link formation was not observed between helix-1 of subunit c and helix-4 of subunit a in 19 additional combinations of doubly Cys-substituted proteins. These results provide direct chemical evidence that helix-2 of subunit c and helix-4 of subunit a pack close enough to each other in the membrane to interact during function. The proximity of helices supports the possibility of an interaction between Arg210 in helix-4 of subunit a and Asp61 in helix-2 of subunit c during proton translocation, as has been suggested previously.

Mutagenesis of beta-Glu-195 of the Rhodospirillum rubrum F1-ATPase and its role in divalent cation-dependent catalysis

We introduced mutations at the fully conserved residue Glu-195 in subunit beta of Rhodospirillum rubrum F1-ATPase. The activities of the expressed wild type (WT) and mutant beta subunits were assayed by following their capacity to assemble into the earlier prepared beta-depleted, membrane-bound R. rubrum enzyme (Philosoph, S., Binder, A., and Gromet-Elhanan, Z. (1977) J. Biol. Chem. 252, 8742-8747) and to restore ATP synthesis and/or hydrolysis activity. All three mutations, beta-E195K, beta-E195Q, and beta-E195G, were found to bind as the WTbeta into the beta-depleted enzyme. They restored between 30 and 60% of the WT restored photophosphorylation activity and 16, 45, and 105%, respectively of the CaATPase activity. The mutants required, however, much higher concentrations of divalent cations and could not restore any significant MgATPase or MnATPase activities. Only beta-E195G could restore some of these activities when assayed in the presence of 100 mM sulfite and high MgCl2 or MnCl2 concentrations. These results suggest that the observed difference in restoration of ATP synthesis and CaATPase, as compared with MgATPase and MnATPase, can be due to the tight regulation of the last two activities, resulting in their inhibition at cation/ATP ratios above 0.5. The R. rubrum F1beta-E195 is equivalent to the mitochondrial F1beta-E199, which points into the tunnel leading to the F1 catalytic nucleotide binding sites (Abrahams, J. P., Leslie, A. G. W., Lutter, R., and Walker, J. E. (1994) Nature 370, 621-628). Our findings indicate that this residue, although not an integral part of the F1 catalytic sites, affects divalent cation binding and release of inhibitory MgADP, suggesting its participation in the interconversion of the F1 catalytic sites between different conformational states.

Three-stepped rotation of subunits gamma and epsilon in single molecules of F-ATPase as revealed by polarized, confocal fluorometry

The proton translocating ATP synthase is conceived as a rotatory molecular engine. ATP hydrolysis by its headpiece, CF1, drives the rotation of subunit gamma relative to the hexagonally arranged large subunits, (alphabeta)3. We investigated transition states of the rotatory drive by polarized confocal fluorometry (POCOF) as applied to single molecules of engineered, immobilized and load-free spinach-CF1. We found that the hydrolysis of ATP caused the stepped and sequential progression of subunit gamma through three discrete angular positions, with the transition states of gamma being too shortlived for detection. We also observed the stepped motion of epsilon, whereas delta was immobile as (alphabeta)3.

Characterization of a b2delta complex from Escherichia coli ATP synthase

The delta subunit of Escherichia coli ATP synthase has been expressed and purified, both as the intact polypeptide and as delta', a proteolytic fragment composed of residues 1-134. The solution structure of delta' as a five-helix bundle has been previously reported (Wilkens, S., Dunn, S. D., Chandler, J., Dahlquist, F. W., and Capaldi, R. A. (1997) Nat. Struct. Biol. 4, 198-201). The delta subunit, in conjunction with delta-depleted F1-ATPase, was fully capable of reconstituting energy-dependent fluorescence quenching in membrane vesicles that had been depleted of F1. A complex of delta with the cytoplasmic domain of the b subunit of F0 was demonstrated and characterized by analytical ultracentrifugation using bST34-156, a form of the b domain lacking aromatic residues. Molecular weight determination by sedimentation equilibrium supported a b2delta subunit stoichiometry. The sedimentation coefficient of the complex, 2.1 S, indicated a frictional ratio of approximately 2, suggesting that delta and the b dimer are arranged in an end-to-end rather than side-by-side manner. These results indicate the feasibility of the b2delta complex reaching from the membrane to the membrane-distal portion of the F1 sector, as required if it is to serve as a second stalk.

Trapping of conformations of the Escherichia coli F1 ATPase by disulfide bond formation. A state of the enzyme with all three catalytic sites of equal and low affinity for nucleotides

A mutant of Escherichia coli F1F0-ATPase, alphaS411C/betaY331W/betaE381C/gammaC87S, has been generated. CuCl2 treatment of this mutant led to cross-linking between alpha and beta subunits in yields of up to 90%. This cross-linking across non-catalytic site interfaces inhibited ATP hydrolysis activity. In the absence of cross-linking, MgATP bound in catalytic sites of the mutant with three different affinities of 0.1 microM, 6 microM and 60 microM, respectively, values that are comparable to wild-type. For MgADP, there was one tight site (0.34 microM) and two sites of lower affinity (each 27 microM), again comparable to wild-type enzyme. After cross-linking all three catalytic sites bound MgATP or MgADP with the same relatively low affinity (approximately 60 microM). Thus cross-linking fixed all three catalytic sites in the same conformation. Trypsin cleavage experiments showed that cross-linking fixed the epsilon subunit in the ATP+EDTA conformation.

Intergenic suppression of the gammaM23K uncoupling mutation in F0F1 ATP synthase by betaGlu-381 substitutions: the role of the beta380DELSEED386 segment in energy coupling

We previously demonstrated that the Escherichia coli F0F1-ATP synthase mutation, gammaM23K, caused increased energy of interaction between gamma- and beta-subunits which was correlated to inefficient coupling between catalysis and transport [Al-Shawi, Ketchum and Nakamoto (1997) J. Biol. Chem. 272, 2300-2306]. Based on these results and the X-ray crystallographic structure of bovine F1-ATPase [Abrahams, Leslie, Lutter and Walker (1994) Nature (London) 370, 621-628] gammaM23K is believed to form an ionized hydrogen bond with betaGlu-381 in the conserved beta380DELSEED386 segment. In this report, we further test the role of gamma-beta-subunit interactions by introducing a series of substitutions for betaGlu-381 and gammaArg-242, the residue which forms a hydrogen bond with betaGlu-381 in the wild-type enzyme. betaE381A, D, and Q were able to restore efficient coupling when co-expressed with gammaM23K. All three mutations reversed the increased transition state thermodynamic parameters for steady state ATP hydrolysis caused by gammaM23K. betaE381K by itself caused inefficient coupling, but opposite from the effect of gammaM23K, the transition state thermodynamic parameters were lower than wild-type. These results suggest that the betaE381K mutation perturbs the gamma-beta-subunit interaction and the local conformation of the beta380DELSEED386 segment in a specific way that disrupts the communication of coupling information between transport and catalysis. betaE381A, L, K, and R, and gammaR242L and E mutations perturbed enzyme assembly and stability to varying degrees. These results provide functional evidence that the beta380DELSEED386 segment and its interactions with the gamma-subunit are involved in the mechanism of coupling.

Interrelation between high and low affinity tentoxin binding sites in chloroplast F1-ATPase revealed by synthetic analogues

Eight synthetic analogues of tentoxin (cyclo-(L-N-MeGlu1-L-Leu2-N-MeDeltaZPhe3-Gly4)) modified in residues 1, 2, and 3 were checked for their ability to inhibit and reactivate the ATPase activity of the activated soluble part of chloroplast ATP synthase. The data were consistent with a model involving two binding sites of different affinities for the toxins. The occupancy of the high affinity site (or tight site) gave rise to an inactive complex, whereas filling both sites (tight + loose) gave rise to a complex of variable activity, dependent on the toxin analogue. Competition experiments between tentoxin and nonreactivating analogues allowed discrimination between the absence of binding and a nonproductive binding to the site of lower affinity (or loose site). The affinity for the loose site was not affected significantly by the modifications of the tentoxin molecule, whereas the affinity for the tight site was found notably changed. Increasing the size of side chain 1 or 2 and introducing a net electrical charge both resulted in a decrease of affinity for the tight site, but the second change dominated the first one. The activity of different ternary complexes enzyme-tentoxin-analogue depended on the nature of the toxin bound on each site and not only on that bound on the loose site. This demonstrates that the reactivation process results from an interaction, direct or not, between these two binding sites. Possible molecular mechanisms are discussed.

Mode of interaction of the single a subunit with the multimeric c subunits during the translocation of the coupling ions by F1F0 ATPases

We have recently isolated a mutant (aK220R, aV264E, aI278N) of the Na+-translocating Escherichia coli/Propionigenium modestum ATPase hybrid with a Na+-inhibited growth phenotype on succinate. ATP hydrolysis by the reconstituted mutant ATPase was inhibited by external (N side) NaCl but not by internal (P side) NaCl. In contrast, LiCl activated the ATPase from the N side and inhibited it from the P side. A similar pattern of activation and inhibition was observed with NaCl and the ATPase from the parent strain PEF42. We conclude from these results that the binding sites for the coupling ions on the c subunits are freely accessible from the N side. Upon occupation of these sites, the ATPase becomes more active, provided that the ions can be further translocated to the P side through a channel of the a subunit. If by mutation of the a subunit this channel becomes impermeable for Na+, N side Na+ ions specifically inhibit the ATPase activity. These conclusions were corroborated by the observation that proton transport into proteoliposomes containing the mutant ATPase was abolished by N side but not by P side Na+ ions. In contrast, LiCl affected proton translocation from either side, similar to the sidedness effect of Na+ ions on H+ transport by the parent hybrid ATPase. If the ATPase carrying the mutated a subunit was incubated with 22NaCl and ATP, 1 mol 22Na+/mol enzyme was occluded. With the parent hybrid ATPase, 22Na+ occlusion was not observed. The occluded 22Na+ could be removed from its tight binding site by 20 mM LiCl, while incubation with 20 mM NaCl was without effect. Li+ but not Na+ is therefore apparently able to pass through the mutated a subunit and make the entrapped Na+ ions accessible again to the aqueous environment. These results suggest an ion translocation mechanism through F0 that in the ATP hydrolysis mode involves binding of the coupling ions from the cytoplasm to the multiple c subunits, ATP-driven rotation to bring a Na+, Li+, or H+-loaded c subunit into a contact site with the a subunit and release of the coupling ions through the a subunit channel to the periplasmic surface of the membrane.

Energy transduction in ATP synthase

Mitochondria, bacteria and chloroplasts use the free energy stored in transmembrane ion gradients to manufacture ATP by the action of ATP synthase. This enzyme consists of two principal domains. The asymmetric membrane-spanning F0 portion contains the proton channel, and the soluble F1 portion contains three catalytic sites which cooperate in the synthetic reactions. The flow of protons through F0 is thought to generate a torque which is transmitted to F1 by an asymmetric shaft, the coiled-coil gamma-subunit. This acts as a rotating 'cam' within F1, sequentially releasing ATPs from the three active sites. The free-energy difference across the inner membrane of mitochondria and bacteria is sufficient to produce three ATPs per twelve protons passing through the motor. It has been suggested that this proton motive force biases the rotor's diffusion so that F0 constitutes a rotary motor turning the gamma shaft. Here we show that biased diffusion, augmented by electrostatic forces, does indeed generate sufficient torque to account for ATP production. Moreover, the motor's reversibility-supplying torque from ATP hydrolysis in F1 converts the motor into an efficient proton pump-can also be explained by our model.

ATP synthesis by F0F1-ATP synthase independent of noncatalytic nucleotide binding sites and insensitive to azide inhibition

ATP hydrolyzing activity of a mutant alpha3beta3gamma subcomplex of F0F1-ATP synthase (DeltaNC) from the thermophilic Bacillus PS3, which lacked noncatalytic nucleotide binding sites, was inactivated completely soon after starting the reaction (Matsui, T., Muneyuki, E. , Honda, M., Allison, W. S., Dou, C., and Yoshida, M. (1997) J. Biol. Chem. 272, 8215-8221). This inactivation is caused by rapid accumulation of the "MgADP inhibited form" which, in the case of wild-type enzyme, would be relieved by ATP binding to noncatalytic sites. We reconstituted F0F1-ATP synthase into liposomes together with bacteriorhodopsin and measured illumination-driven ATP synthesis. Remarkably, DeltaNC F0F1-ATP synthase catalyzed continuous turnover of ATP synthesis while it could not promote ATP-driven proton translocation. ATP synthesis by DeltaNC F0F1-ATP synthase, as well as wild-type enzyme, proceeded even in the presence of azide, an inhibitor of ATP hydrolysis that stabilizes the MgADP inhibited form. The time course of ATP synthesis by DeltaNC F0F1-ATP synthase was linear, and gradual acceleration to the maximal rate, which was observed for the wild-type enzyme, was not seen. Thus, ATP synthesis can proceed without nucleotide binding to noncatalytic sites even though the rate is sub-maximal. These results indicate that the MgADP inhibited form is not produced in ATP synthesis reaction, and in this regard, ATP synthesis may not be a simple reversal of ATP hydrolysis.

The ATP synthase atpHAGDC (F1) operon from Rhodobacter capsulatus

The atpHAGDC operon of Rhodobacter capsulatus, containing the five genes coding for the F1 sector of the ATP synthase, has been cloned and sequenced. The promoter region has been defined by primer extension analysis. It was not possible to obtain viable cells carrying atp deletions in the R. capsulatus chromosome, indicating that genes coding for ATP synthase are essential, at least under the growth conditions tested. We were able to circumvent this problem by combining gene transfer agent transduction with conjugation. This method represents an easy way to construct strains carrying mutations in indispensable genes.

Visualization of a peripheral stalk in V-type ATPase: evidence for the stator structure essential to rotational catalysis

F- and V-type ATPases are central enzymes in energy metabolism that couple synthesis or hydrolysis of ATP to the translocation of H+ or Na+ across biological membranes. They consist of a soluble headpiece that contains the catalytic sites and an integral membrane-bound part that conducts the ion flow. Energy coupling is thought to occur through the physical rotation of a stalk that connects the two parts of the enzyme complex. This mechanism implies that a stator-like structure prevents the rotation of the headpiece relative to the membrane-bound part. Such a structure has not been observed to date. Here, we report the projected structure of the V-type Na+-ATPase of Clostridium fervidus as determined by electron microscopy. Besides the central stalk, a second stalk of 130 A in length is observed that connects the headpiece and membrane-bound part in the periphery of the complex. This additional stalk is likely to be the stator.

Nucleotide-depleted beef heart F1-ATPase exhibits strong positive catalytic cooperativity

Catalytic cooperativity is a central feature of the binding change mechanism for F0F1-ATP synthases. However, in a recent publication (Reynafarje, B. D., and Pedersen, P. L. (1996) J. Biol. Chem. 271, 32546-32550), Reynafarje and Pedersen claim that cooperative effects are an artifact caused by endogenous nucleotides and that when such nucleotides are removed, the multiple catalytic sites on MF1 behave independently during ATP hydrolysis. In contrast to this conclusion, we show here that when ATP is loaded at a single catalytic site on nucleotide-depleted MF1, the rate of product release is accelerated by up to 5 x 10(4)-fold by the binding of ATP at adjacent catalytic sites. Hence, nucleotide-depleted MF1 is not an exception but does in fact show strong cooperative interactions. In addition, evidence is presented supporting a random order for product release during ATP hydrolysis.

Subunit interactions of Escherichia coli F1-ATPase: mutants of the gamma subunits defective in interaction with the epsilon subunit isolated by the yeast two-hybrid system

Previously, we established a method to detect subunit interactions of F1-ATPase by the yeast two-hybrid system (Moritani, C., et al. Biochim. Biophys. Acta 1274, 67-72, 1996). Here, we isolated mutants of the gamma subunits defective in interaction with the epsilon subunit by this new procedure to study the molecular basis of coupling mechanisms of the F1F0-ATPase. Based on the intensities of the reporter gene expression in this system, five mutants of the gamma subunit with different levels of gamma-epsilon interactions were isolated and their single base substitutions were determined. Mutants with a substitution of Pro-55 for Leu, Thr-102 for Met, Val-141 for Asp, or Gln-235 for Leu exhibited decreased reporter gene expression, suggesting decreased levels of interaction, while Asp-85 for Gly mutation caused a higher level of expression, suggesting increased interaction. Among these point mutations, G85D, M102T, or D141V mutations were introduced into the gamma subunit gene in the plasmid carrying whole unc operon. Transformants carrying a deletion mutant of the whole unc operon with these expression plasmids were analyzed. Mutations M102T and D141V with decreased gamma-epsilon interaction caused increases of membrane-bound F1-ATPase activity and proton pumping activity, while G85D with increased gamma-epsilon interaction exhibited lower levels of F1-ATPase activity in the membranes. Molecular assembly of the F1 subunits on the mutant membranes detected by Western blotting exhibited no defect for all three mutants. These results suggested that the correlation between the ATPase activity and gamma-epsilon interaction is reciprocal and this interaction may regulate the ATPase activity. The topological and functional importance of Gly-85, Met-102, and Asp-141 together with Leu-55 and Leu-235 in gamma-epsilon interaction is discussed.

Interaction of the delta and b subunits contributes to F1 and F0 interaction in the Escherichia coli F1F0-ATPase

Interactions of the F1F0-ATPase subunits between the cytoplasmic domain of the b subunit (residues 26-156, bcyt) and other membrane peripheral subunits including alpha, beta, gamma, delta, epsilon, and putative cytoplasmic domains of the a subunit were analyzed with the yeast two-hybrid system and in vitro reconstitution of ATPase from the purified subunits as well. Only the combination of bcyt fused to the activation domain of the yeast GAL-4, and delta subunit fused to the DNA binding domain resulted in the strong expression of the beta-galactosidase reporter gene, suggesting a specific interaction of these subunits. Expression of bcyt fused to glutathione S-transferase (GST) together with the delta subunit in Escherichia coli resulted in the overproduction of these subunits in soluble form, whereas expression of the GST-bcyt fusion alone had no such effect, indicating that GST-bcyt was protected by the co-expressed delta subunit from proteolytic attack in the cell. These results indicated that the membrane peripheral domain of b subunit stably interacted with the delta subunit in the cell. The affinity purified GST-bcyt did not contain significant amounts of delta, suggesting that the interaction of these subunits was relatively weak. Binding of these subunits observed in a direct binding assay significantly supported the capability of binding of the subunits. The ATPase activity was reconstituted from the purified bcyt together with alpha, beta, gamma, delta, and epsilon, or with the same combination except epsilon. Specific elution of the ATPase activity from glutathione affinity column with the addition of glutathione after reconstitution demonstrated that the reconstituted ATPase formed a complex. The result indicated that interaction of b and delta was stabilized by F1 subunits other than epsilon and also suggested that b-delta interaction was important for F1-F0 interaction.

Intramolecular rotation in ATP synthase: dynamic and crystallographic studies on thermophilic F1

A single molecule of ATP synthase (F0F1) is by itself a rotary motor, the smallest ever found, and this biomotor is driven by an electrochemical potential of H+ (delta microH+). F0F1 is composed of an ion-conducting portion (F0) and a catalytic portion (F1). The major breakthroughs in studies on the mechanochemical coupling have been the direct observation of the rotation of a stable alpha 3 beta 3 gamma complex of thermophilic F1 (TF1), and X-ray crystallography of the alpha 3 beta 3 gamma portion of mitochondrial F1 (MF1) and the alpha 3 beta 3 oligomer of TF1. This review focuses on the dynamics of TF1, demonstrated by a crucial experiment. The torque of the rotation was estimated to be 42 pN.nm from the delta microH+ and frictional force. Important unsolved problems are the crystallography of F0, elastic energy conversion, and the stator and rotor of this biomotor.

ATP synthase: an electrochemical transducer with rotatory mechanics

ATP synthase (F0F1-ATPase) uses proton- or sodium-motive force to produce ATP form ADP and P(i). Three lines of experiment have recently demonstrated large-scale intersubunit rotation during ATP hydrolysis by F1. We discuss how ion flow through the membrane-intrinsic portion, F0, may generate torque and how this might be transmitted between stator and rotor to finally expel spontaneously formed ATP from F1 into water.

Acceleration of unisite catalysis of mitochondrial F1-adenosinetriphosphatase by ATP, ADP and pyrophosphate--hydrolysis and release of the previously bound [gamma-32P]ATP

The effect of ATP, ADP and pyrophosphate (PPi) on hydrolysis and release of [gamma-32P]ATP bound to the high-affinity catalytic site of soluble F1 from bovine heart mitochondria under unisite conditions [Grubmeyer, C., Cross, R. L. & Penefsky, H. S. (1982) J. Biol. Chem. 257, 12092-12100] was studied. In accord with the previous data, it was observed that millimolar concentrations of ATP or ADP added to F1 undergoing unisite hydrolysis of [gamma-32P]ATP accelerated its hydrolysis. PPi also produced a hydrolytic burst of a fraction of the previously bound [gamma-32P]ATP; kinetic data suggested that for production of optimal hydrolysis by PPi of the bound [gamma-32P]ATP, two binding sites with apparent Kd of 27 microM and 240 microM must be filled. The extent of the hydrolytic burst induced by MgPPi was lower than that induced by ADP and ATP. In F1 in which PPi had produced a hydrolytic burst of the bound [gamma-32P]ATP, the addition of ATP induced a second burst of hydrolysis. By filtration experiments and enzyme trapping, it was also studied whether ATP, ADP and PPi produce release of the tightly bound [gamma-32P]ATP. At millimolar concentrations, ATP and ADP brought about release of about 25% of the previously bound [gamma-32P]ATP. At micromolar concentrations, ADP accelerated the hydrolysis of the previously bound [gamma-32P]ATP but not its release. Hence, the hydrolytic and release reactions could be separated, indicating that the two reactions require the occupancy of different sites in F1. With PPi, no release of the tightly bound [gamma-32P]ATP was observed. The ADP induced hydrolysis and release of the F1-bound [gamma-32P]ATP were inhibited by sodium azide to the same extent (60%). Since release of ATP from a high-affinity catalytic site of F1 represents the terminal step of oxidative phosphorylation, the data illustrate that the binding energy of substrates to F1 is critical to the ejection of ATP into the media. The failure of PPi to induce release of [gamma-32P]ATP bound to F1 under unisite conditions is probably due to its lower binding energy.

Thermophilic F1-ATPase is activated without dissociation of an endogenous inhibitor, epsilon subunit

Subunit complexes (alpha3beta3gamma, alpha3beta3gammadelta, alpha3beta3gammaepsilon, and alpha3beta3gammadeltaepsilon) of thermophilic F1-ATPase were prepared, and their catalytic properties were compared to know the role of delta and epsilon subunits in catalysis. The presence of delta subunit in the complexes had slight inhibitory effect on the ATPase activity. The effect of epsilon subunit was more profound. The (-epsilon) complexes, alpha3beta3gamma and alpha3beta3gammadelta, initiated ATP hydrolysis without a lag. In contrast, the (+epsilon) complexes, alpha3beta3gammaepsilon and alpha3beta3gammadeltaepsilon, started hydrolysis of ATP (<700 microM) with a lag phase that was gradually activated during catalytic turnover. As ATP concentration increased, the lag phase of the (+epsilon) complexes became shorter, and it was not observed above 1 mM ATP. Analysis of binding and hydrolysis of the ATP analog, 2',3'-O-(2,4,6-trinitrophenyl)-ATP, suggested that the (+epsilon) complexes bound substrate only slowly. Differing from Escherichia coli F1-ATPase, the activation of the (+epsilon) complexes from the lag phase was not due to dissociation of epsilon subunit since the re-isolated activated complex retained epsilon subunit. This indicates that there are two alternative forms of the (+epsilon) complex, inhibited form and activated form, and the inhibited one is converted to the activated one during catalytic turnover.

Structure and arrangement of the delta subunit in the E. coli ATP synthase (ECF1F0)

F1F0 type ATPases are made up of two parts, an F1, which contains three catalytic sites on beta subunits, and an F0 which contains the proton channel. These two domains have been visualized in electron microscopy as linked by a narrow stalk of around 45 A in length. Biochemical studies have provided clear evidence that the gamma and epsilon subunits are components of this stalk. There is an emerging consensus that the gamma and epsilon subunits rotate relative to the alpha 3 beta 3 domain as part of the cooperativity and energy coupling within the complex. Two other subunits are required to link the F1 to F0 in the E. coli enzyme, and these are the delta and b subunits. The structure of a major part of the delta subunit (residues 1-134) has now been obtained by NMR spectroscopy. The main feature is a six alpha-helix bundle, which provides the N-terminal domain of the delta subunit. This domain interacts with the F1 core via the N-terminal part of the alpha subunit. The C-terminal domain of delta is less well defined. This part is required for binding to the F0 part by direct interaction with the b subunits. It is argued that delta and the two copies of the b subunit are components of a second stalk linking the F1 and F0 parts, which acts as a stator to allow the energy-linked rotational movements of delta and epsilon subunits.

Cross-linking of chloroplast F0F1-ATPase subunit epsilon to gamma without effect on activity. Epsilon and gamma are parts of the rotor

Cys residues were directed into positions 17, 28, 41 and 85 of a Cys6-->Ser mutant of subunit epsilon of spinach chloroplast F0F1 ATP synthase. Wild-type and engineered epsilon were expressed in Escherichia coli, purified in the presence of urea, refolded and reassembled with spinach chloroplast F1 lacking the epsilon subunit [F1(-epsilon)]. Cys-containing epsilon variants were modified with a sulfhydryl-reactive photolabile cross-linker. Photocross-linking of epsilon to F1(-epsilon) yielded the same SDS gel pattern of cross-link products independent of the presence or absence of Mg2+ x ADP, phosphate and Mg2+ x ATP. Epsilon (wild type) [Ser6,Cys28]epsilon and [Ser6,Cys41]epsilon were cross-linked with subunit gamma. With chloroplast F0F1 the same cross-link pattern was obtained, except for one extra cross-link, probably between [Ser6,Cys28]epsilon and F0 subunit III. [Ser6,Cys17]epsilon and [Ser6,Cys85]epsilon did not produce cross-links. Cross-linking of epsilon, [Ser6,Cys28]epsilon, [Ser6,Cys41]epsilon to gamma in soluble chloroplast F1 impaired the ability of epsilon to inhibit Ca2+-ATPase activity. The Mg2+-ATPase activity of soluble F1 (measured in the presence of 30% MeOH) was not affected by cross-linking epsilon with gamma. Functional reconstitution of photophosphorylation in F1-depleted thylakoids was observed with F1 in which gamma was cross-linked to [Ser6,Cys28]epsilon or [Ser6,Cys41]epsilon but not with wild-type epsilon. In view of the intersubunit rotation of gamma relative to (alphabeta)3, which is driven by ATP hydrolysis, gamma and epsilon would seem to act concertedly as parts of the 'rotor' relative to the 'stator' (alphabeta)3.

Subunit rotation in Escherichia coli FoF1-ATP synthase during oxidative phosphorylation

We report evidence for proton-driven subunit rotation in membrane-bound FoF1-ATP synthase during oxidative phosphorylation. A betaD380C/gammaC87 crosslinked hybrid F1 having epitope-tagged betaD380C subunits (betaflag) exclusively in the two noncrosslinked positions was bound to Fo in F1-depleted membranes. After reduction of the beta-gamma crosslink, a brief exposure to conditions for ATP synthesis followed by reoxidation resulted in a significant amount of betaflag appearing in the beta-gamma crosslinked product. Such a reorientation of gammaC87 relative to the three beta subunits can only occur through subunit rotation. Rotation was inhibited when proton transport through Fo was blocked or when ADP and Pi were omitted. These results establish FoF1 as the second example in nature where proton transport is coupled to subunit rotation.

ATP synthase: a tentative structural model

Adenosine triphosphate (ATP) synthase produces ATP from ADP and inorganic phosphate at the expense of proton- or sodium-motive force across the respective coupling membrane in Archaea, Bacteria and Eucarya. Cation flow through the intrinsic membrane portion of this enzyme (Fo, subunits ab2c9-12) and substrate turnover in the headpiece (F1, subunits alpha3beta3 gammadeltaepsilon) are mechanically coupled by the rotation of subunit gamma in the center of the catalytic hexagon of subunits (alphabeta)3 in F1. ATP synthase is the smallest rotatory engine in nature. With respect to the headpiece alone, it probably operates with three steps. Partial structures of six out of its at least eight different subunits have been published and a 3-dimensional structure is available for the assembly (alphabeta)3gamma. In this article, we review the available structural data and build a tentative topological model of the holoenzyme. The rotor portion is proposed to consist of a wheel of at least nine copies of subunits c, epsilon and a portion of gamma as a spoke, and another portion of gamma as a crankshaft. The stator is made up from a, the transmembrane portion of b2, delta and the catalytic hexagon of (alphabeta)3. As an educated guess, the model may be of heuristic value for ongoing studies on this fascinating electrochemical-to-mechanical-to-chemical transducer.

Sulfite stimulates the ATP hydrolysis activity of but not proton translocation by the ATP synthase of Rhodobacter capsulatus and interferes with its activation by delta muH+

Sulfite stimulates the rate of ATP hydrolysis by the ATP synthase in chromatophores of Rhodobacter capsulatus. The stimulated activity is inhibited by oligomycin. The activation takes place also in uncoupled chromatophores. The activation consists in an increase of about 12-15-fold of the Vmax for the ATP hydrolysis reaction, while the Km for MgATP is unaffected at 0.16+/-0.03 mM. The dependence of Vmax on the sulfite concentration follows a hyperbolic pattern with half maximum effect at 12 mM. Sulfite affects the ability of the enzyme in translocating protons. Concomitant measurements of the rate of ATP hydrolysis and of ATP-induced protonic flows demonstrate that at sulfite concentrations of greater than 10 mM the hydrolytic reaction becomes progressively uncoupled from the process of proton translocation. This is accompanied by an inhibition of ATP synthesis, either driven by light or by artificially induced ionic gradients. ATP synthesis is totally inhibited at concentrations of at least 80 mM. Sulfite interferes with the mechanism of activation by delta muH+. Low concentrations of this anion (< or = 2 mM) prevent the activation by delta muH+. At higher concentrations a marked stimulation of the activity prevails, regardless of the occurrence of a delta muH+ across the membrane. Phosphate at millimolar concentrations can reverse the inhibition by sulfite. These experimental results can be simulated by a model assuming multiple and competitive equilibria for phosphate or sulfite binding with two binding sites for the two ligands (for sulfite K1S = 0.26 and K2S = 37 mM, and for phosphate K1P = 0.06 and K2P = 4.22 mM), and in which the state bound only to one sulfite molecule is totally inactive in hydrolysis. The competition between phosphate and sulfite is consistent with the molecular structures of the two ligands and of the enzyme.

The regulatory functions of the gamma and epsilon subunits from chloroplast CF1 are transferred to the core complex, alpha3beta3, from thermophilic bacterial F1

The expression plasmids for the subunit gamma (gamma(c)) and the subunit epsilon (epsilon(c)) of chloroplast coupling factor (CF1) from spinach were constructed, and the desired proteins were expressed in Escherichia coli. Both expressed subunits were obtained as inclusion bodies. When recombinant gamma(c) was mixed with recombinant alpha and beta subunits of F1 from thermophilic Bacillus PS3 (TF1), a chimeric subunit complex (alpha3beta3gamma(c)) was reconstituted and it showed significant ATP hydrolysis activity. The ATP hydrolysis activity of this complex was enhanced in the presence of dithiothreitol and suppressed by the addition of CuCl2, which induces formation of a disulfide bond between two cysteine residues in gamma(c). Hence, this complex has similar modulation characteristics as CF1. The effects of recombinant epsilon(c) and epsilon subunit from TF1 (epsilon(t)) on alpha3beta3gamma(c) were also investigated. Epsilon(c) strongly inhibited the ATP hydrolysis activity of chimeric alpha3beta3gamma(c) complex but epsilon(t) did not. The inhibition was abolished and the ATP hydrolysis activity was recovered when methanol was added to the assay medium. The addition of epsilon(c) or epsilon(t) to the alpha3beta3gamma(t) complex, which is the authentic subunit complex from TF1, resulted in weak stimulation of the ATP hydrolysis activity. These results suggest that (a) the specific regulatory function of gamma(c) can be transferred to the bacterial subunit complex; (b) the interaction between the gamma(c) subunit and epsilon(c) strongly affects the enzyme activity, which was catalyzed at the catalytic sites that reside on the alpha3beta3 core.

Rotation of a gamma-epsilon subunit domain in the Escherichia coli F1F0-ATP synthase complex. The gamma-epsilon subunits are essentially randomly distributed relative to the alpha3beta3delta domain in the intact complex

A triple mutant of Escherichia coli F1F0-ATP synthase, alphaQ2C/alphaS411C/epsilonS108C, has been generated for studying movements of the gamma and epsilon subunits during functioning of the enzyme. It includes mutations that allow disulfide bond formation between the Cys at alpha411 and both Cys-87 of gamma and Cys-108 of epsilon, two covalent cross-links that block enzyme function (Aggeler, R., and Capaldi, R. A. (1996) J. Biol. Chem. 271, 13888-13891). A cross-link is also generated between the Cys at alpha2 and Cys-140 of the delta subunit, which has no effect on functioning (Ogilvie, I., Aggeler, R., and Capaldi, R. A. (1997) J. Biol. Chem. 272, 16652-16656). CuCl2 treatment of the mutant alphaQ2C/alphaS411C/epsilonS108C generated five major cross-linked products. These are alpha-gamma-delta, alpha-gamma, alpha-delta-epsilon, alpha-delta, and alpha-epsilon. The ratio of alpha-gamma-delta to the alpha-gamma product was close to 1:2, i.e. in one-third of the ECF1F0 molecules the gamma subunit was attached to the alpha subunit at which the delta subunit is bound. Also, 20% of the epsilon subunit was present as a alpha-delta-epsilon product. With regard to the delta subunit, 30% was in the alpha-gamma-delta, 20% in the alpha-delta-epsilon, and 50% in the alpha-delta products when the cross-linking was done after incubation in ATP + MgCl2. The amounts of these three products were 40, 22, and 38%, respectively, in experiments where Cu2+ was added after preincubation in ATP + Mg2+ + azide. The delta subunit is fixed to, and therefore identifies, one specific alpha subunit (alphadelta). A distribution of the gamma and epsilon subunits, which is essentially random with respect to the alpha subunits, can only be explained by rotation of gamma-epsilon relative to the alpha3beta3 domain in ECF1F0.

Subunit c from the sodium-ion-translocating F1F0-ATPase of Propionigenium modestum--production, purification and properties of the protein in dodecylsulfate solution

Escherichia coli strain PEF42 produces a sodium-ion-dependent hybrid F1F0-ATPase consisting of the Propionigenium modestum subunits a, b, c and delta, of a hybrid alpha subunit and of the E. coli subunits beta, gamma and epsilon. The gene encoding subunit c of the P. modestum F1F0-ATPase was cloned into the pT7-7 expression vector to yield plasmid pT7c. E. coli PEF42 was transformed with plasmid pT7c together with plasmid pGP1-2, which harbours the gene for the T7 RNA polymerase. The production of the P. modestum subunit c was induced by a temperature shift from 30 degrees C to 42 degrees C for 30 min and led to an increased concentration of this protein in the membrane of the host strain. The c subunit produced in E. coli moved as a monomer in dodecylsulfate electrophoresis. The protein was extracted from the cells with chloroform/methanol, purified and incorporated into sodium dodecylsulfate micelles. Circular dichroism of subunit c in sodium dodecylsulfate showed a temperature-stable spectrum (between 20-60 degrees C) with a high proportion of alpah-helical structure. Upon incubation of subunit c with [14C]dicyclohexylcarbodiimide the protein became labelled in a sodium-ion-dependent manner, similar to the labelling observed if the purified F1F0-ATPase of P. modestum, was treated with the radioactive carbodiimide. The Na+-specific site was therefore retained in the isolated c subunit dissolved in dodecylsulfate.

Mutations in subunit 6 of the F1F0-ATP synthase cause two entirely different diseases

A lowered efficiency of oxidative phosphorylation was recently found in a Leber hereditary optic neuropathy (LHON) proband carrying a mutation in the mtDNA gene for subunit 6 of the membrane-bound F0 segment of the F1F0-ATP synthase [9]. This phenotype was transferred to cytoplasmic hybrid cells together with the mutation, proving its functional significance. Increasing the respiratory rate in the mitochondria from this mutant raised the ATP/2e- ratio back to normal values. A different mutation in the same mtDNA gene has been found in patients with the NARP syndrome [10]. Although the ATP/2e- ratio is also decreased in this mutant, in this case an increase in the respiratory rate could not compensate for it. Whilst both mutations affect subunit 6 of the proton-translocating F0 segment, the LHON mutation induces a proton leak whereas the NARP mutation blocks proton translocation. Hence, the latter will have much more destructive metabolic consequences in agreement with the large clinical differences between the two diseases.

Characterization of the alpha and beta-subunits of the F0F1-ATPase from the alga Polytomella spp., a colorless relative of Chlamydomonas reinhardtii

The isolation and partial characterization of the oligomycin-sensitive F0F1-ATP synthase/ATPase from the colorless alga Polytomella spp. is described. Purification was performed by solubilization with dodecyl-beta-D-maltoside followed by Sepharose Hexyl ammonium chromatography, a matrix that interacts with the F1 sector of mitochondrial ATPases. The alpha-subunit, which migrates on SDS-polyacrylamide gels with an apparent molecular mass of 55 kDa, was identified by the N-terminal sequencing of 47 residues. This subunit exhibited a short extension at its N-terminus highly similar to the one described for the unicellular alga Chlamydomonas reinhardtii (Nurani, G. and Franzén L.-G. (1996) Plant Mol. Biol. 31, 1105-1116). In whole mitochondria, the alpha-subunit was susceptible to limited proteolytic digestion induced by heat. An endogenous protease removed the first 22 residues of the mature alpha-subunit. Subunit beta was also identified by N-terminal sequencing of 31 residues. This subunit of 63 kDa exhibited a higher apparent molecular mass than alpha, as judged by its mobility on denaturing polyacrylamide gel electrophoresis. This beta-subunit is 7-8 kDa larger than the beta-subunits of other mitochondrial ATPases. It is suggested that the beta-subunit from Polytomella spp. may have a C-terminal extension similar to that described for the green alga C. reinhardtii (Franzén, L.-G. and Falk, G.(1992) Plant Mol. Biol. 19, 771-780). In addition, it was found that the C-terminal extension of the beta-subunit of C. reinhardtii showed homology with the endogenous ATPase inhibitors from various sources and with the epsilon-subunit from the F0F1-ATP synthase from Escherichia coli, which is considered to be a functional homolog of the inhibitor proteins. The data reported here provide the first biochemical evidence for a close relationship between the colorless alga Polytomella spp. and its photosynthetic counterpart C. reinhardtii. It is also suggested that the C-terminal extensions of the beta-subunits of the ATP synthases from these algae, may play a regulatory role in these enzymes.

Chloroplast F0F1 ATP Synthase Imaged by Atomic Force Microscopy

The F0F1 ATP synthase of chloroplasts was imaged using atomic force microscopy (AFM) in contact mode under physiological conditions. Chloroplast (CF0F1) ATP synthases were reconstituted into liposomes. Liposomes were adsorbed on a mica surface where they spread and formed lipid bilayers containing CF0F1 ATP synthases which could be imaged. From these reconstituted CF0F1 ATP synthases, the CF1 part could be removed either by application of a chemical denaturant or less efficiently by mechanical stripping with the AFM tip. Embedded in the lipid bilayer were seen ring-like structures with a central dimple with outer diameters of 20 +/- 3 nm (chemical denaturant) and ca. 7 nm (mechanical stripping), respectively. Ring-like structures were also observed in a protein-free lipid bilayer. These had diameters of 30 +/- 5 nm and could be clearly distinguished from the structures observed after mechanical stripping. Hence, the ring-like structures observed after mechanical stripping might represent the intrinsic membrane domain CF0 or the oligomer of its subunit III. In addition, isolated CF1 adsorbed directly onto the mica surface was imaged. In accordance with the size known from electron microscopy, a diameter of 13 +/- 4 nm was measured.

Model of the c-subunit oligomer in the membrane domain of F-ATPases

A model is described of a dodecameric complex consisting of the integral membrane component subunit c of the H+-transporting Fo domain of Escherichia coli F-ATPase. A high-resolution partial structure of monomeric subunit c resulting from 1H-NMR studies [1] was used for constructing the model. The validity of the proposed arrangement of protomers in the dodecameric complex was tested by amino acid substitution analysis and chemical, biochemical and genetic data on subunit c.

Cross-linking of the delta subunit to one of the three alpha subunits has no effect on functioning, as expected if delta is a part of the stator that links the F1 and F0 parts of the Escherichia coli ATP synthase

A mutant of the Escherichia coli F1F0-ATPase has been generated (alphaQ2C) in which the glutamine at position 2 of the alpha subunit has been replaced with a cysteine residue. Cu2+ treatment of ECF1 from this mutant cross-linked an alpha subunit to the delta subunit in high yield. Two different sites of disulfide bond formation were involved, i.e. between Cys90 (or the closely spaced Cys47) of alpha with Cys140 of delta, and between Cys2 of alpha and Cys140 of delta. Small amounts of other cross-linked products, including alpha-alpha, delta internal, and alpha-alpha-delta were obtained. In ECF1F0, there was no cross-linking between the intrinsic Cys of alpha and Cys140. Instead, the product generated between Cys2 of alpha and Cys140 of delta was obtained at near 90% yield. Small amounts of alpha-alpha and delta internal were present, and under high Cu2+ concentrations, alpha-alpha-delta was also formed. The ATPase activity of ECF1 and ECF1F0 was not significantly affected by the presence of these cross-links. When Cys140 of delta was first modified with N-ethylmaleimide in ECF1F0, an alpha-delta cross-link was still produced, although in lower yield, between Cys64 of delta and Cys2 of alpha. ATP hydrolysis-linked proton pumping of inner membranes from the mutant alpha2QC was only marginally affected by cross-linking of the alpha to the delta subunit. These results indicate that Cys140 and Cys64 of the delta subunit and Cys2 of the alpha subunit are in close proximity. This places the delta subunit near the top of the alpha-beta hexagon and not in the stalk region. As fixing the delta to the alpha by cross-linking does not greatly impair either the ATPase function of the enzyme, or coupled proton translocation, we argue that the delta subunit forms a portion of the stator linking F1 to F0.

The vacuolar H+-ATPase: a universal proton pump of eukaryotes

The vacuolar H+-ATPase (V-ATPase) is a universal component of eukaryotic organisms. It is present in the membranes of many organelles, where its proton-pumping action creates the low intra-vacuolar pH found, for example, in lysosomes. In addition, there are a number of differentiated cell types that have V-ATPases on their surface that contribute to the physiological functions of these cells. The V-ATPase is a multi-subunit enzyme composed of a membrane sector and a cytosolic catalytic sector. It is related to the familiar FoF1 ATP synthase (F-ATPase), having the same basic architectural construction, and many of the subunits from the two display identity with one another. All the core subunits of the V-ATPase have now been identified and much is known about the assembly, regulation and pharmacology of the enzyme. Recent genetic analysis has shown the V-ATPase to be a vital component of higher eukaryotes. At least one of the subunits, i.e. subunit c (ductin), may have multifunctional roles in membrane transport, providing a possible pathway of communication between cells. The structure of the membrane sector is known in some detail, and it is possible to begin to suggest how proton pumping is coupled to ATP hydrolysis.

The crystal structure of the nucleotide-free alpha 3 beta 3 subcomplex of F1-ATPase from the thermophilic Bacillus PS3 is a symmetric trimer

BACKGROUND

F1-ATPase, an oligomeric assembly with subunit stoichiometry alpha 3 beta 3 gamma delta epsilon, is the catalytic component of the ATP synthase complex, which plays a central role in energy transduction in bacteria, chloroplasts and mitochondria. The crystal structure of bovine mitochondrial F1-ATPase displays a marked asymmetry in the conformation and nucleotide content of the catalytic beta subunits. The alpha 3 beta 3 subcomplex of F1-ATPase has been assembled from subunits of the moderately thermophilic Bacillus PS3 made in Escherichia coli, and the subcomplex is active but does not show the catalytic cooperativity of intact F1-ATPase. The structure of this subcomplex should provide new information on the conformational variability of F1-ATPase and may provide insights into the unusual catalytic mechanism employed by this enzyme.

RESULTS

The crystal structure of the nucleotide-free bacterial alpha 3 beta 3 subcomplex of F1-ATPase, determined at 3.2 A resolution, shows that the oligomer has exact threefold symmetry. The bacterial beta subunits adopt a conformation essentially identical to that of the nucleotide-free beta subunit in mitochondrial F1-ATPase; the alpha subunits have similar conformations in both structures.

CONCLUSIONS

The structures of the bacterial F1-ATPase alpha and beta subunits are very similar to their counterparts in the mitochondrial enzyme, suggesting a common catalytic mechanism. The study presented here allows an analysis of the different conformations adopted by the alpha and beta subunits and may ultimately further our understanding of this mechanism.

Interactions between the F1 and F0 parts in the Escherichia coli ATP synthase. Associations involving the loop region of C subunits

The N-ethylmaleimide reactivity of c subunits in Escherichia coli F1F0 ATP synthase (ECF1F0) isolated from five mutants, each with a cysteine at a different position in the polar loop region (positions 39, 40, 42, 43, and 44), has been investigated. The maleimide was found to react with Cys placed at positions 42, 43, and 44 but not at 39 or 40. All copies of the c subunit reacted similarly when the Cys was at position 43 or 44. In contrast, the Cys in the mutant cQ42C reacted as two classes, with 60% reacting relatively rapidly and 40% reacting at a rate 40-fold slower. After removing F1, all copies of the c subunit in this mutant reacted equally fast. Therefore, the slow class in the cQ42C mutant represents c subunits shielded by, and probably involved directly in, the interaction of the F0 with gamma and epsilon subunits of the F1 part. Based on the estimated stoichiometry of c subunits in the ECF1F0 complex, 4 or 5 c subunits are involved in this F1 interaction. N-Ethylmaleimide modification of all of the c subunits reduced ATPase activity by only 30% in ECF1F0 from mutant cQ42C. Modification of the more rapidly reacting class had little effect on ATP hydrolysis-driven proton translocation, and did not alter the DCCD inhibition of ATPase activity. However, as those c subunits involved in the F1 interaction became modified, DCCD inhibition was progressively lost, as was coupling between ATP hydrolysis and proton translocation.

Rotation of the gamma subunit in F1-ATPase; evidence that ATP synthase is a rotary motor enzyme

ATP-dependent, azide-sensitive rotation of the gamma subunit relative to the alpha3beta3 hexagonal ring of ATP synthase was observed with a single molecule imaging system. Thus, ATP synthase is a rotary motor enzyme, the first ever found.

The dTTPase mechanism of T7 DNA helicase resembles the binding change mechanism of the F1-ATPase

Bacteriophage T7 DNA helicase is a ring-shaped hexamer that catalyzes duplex DNA unwinding using dTTP hydrolysis as an energy source. Of the six potential nucleotide binding sites on the hexamer, we have found that three are noncatalytic sites and three are catalytic sites. The noncatalytic sites bind nucleotides with a high affinity, but dTTPs bound to these sites do not dissociate or hydrolyze through many dTTPase turnovers at the catalytic sites. The catalytic sites show strong cooperativity which leads to sequential binding and hydrolysis of dTTP. The elucidated dTTPase mechanism of the catalytic sites of T7 helicase is remarkably similar to the binding change mechanism of the ATP synthase. Based on the similarity, a general mechanism for hexameric helicases is proposed. In this mechanism, an F1-ATPase-like rotational movement around the single-stranded DNA, which is bound through the central hole of the hexamer, is proposed to lead to unidirectional translocation along single-stranded DNA and duplex DNA unwinding.

Imaging single-molecule dichroism

Video-enhanced optical fluorescence microscopy has been used to determine the direction of the transition dipole moment of individual molecules. Fast electro-optical switching of the polarization of the excitation allowed us to detect the linear dichroism of single fluorophores in solid phospholipid membranes. Quantitative analysis of the fluorescence signal showed that rotational mobility is absent in solid biomembranes.

Functional and idling rotatory motion within F1-ATPase

ATP synthase mediates proton flow through its membrane portion, F0, which drives the synthesis of ATP in its headpiece, F1. The F1-portion contains a hexagonal array of three subunits alpha and three beta encircling a central subunit gamma, that in turn interacts with a smaller epsilon and with F0. Recently we reported that the application of polarized absorption recovery after photobleaching showed the ATP-driven rotation of gamma over at least two, if not three, beta. Here we extend probes of such rotation aided by a new theory for assessing continuous versus stepped, Brownian versus unidirectional molecular motion. The observed relaxation of the absorption anisotropy is fully compatible with a unidirectional and stepping rotation of gamma over three equidistantly spaced angular positions in the hexagon formed by the alternating subunits alpha and beta. The results strongly support a rotational catalysis with equal participation of all three catalytic sites. In addition we report a limited rotation of gamma without added nucleotides, perhaps idling and of Brownian nature, that covers only a narrow angular domain.

Direct observation of the rotation of F1-ATPase

Cells employ a variety of linear motors, such as myosin, kinesin and RNA polymerase, which move along and exert force on a filamentous structure. But only one rotary motor has been investigated in detail, the bacterial flagellum (a complex of about 100 protein molecules). We now show that a single molecule of F1-ATPase acts as a rotary motor, the smallest known, by direct observation of its motion. A central rotor of radius approximately 1 nm, formed by its gamma-subunit, turns in a stator barrel of radius approximately 5nm formed by three alpha- and three beta-subunits. F1-ATPase, together with the membrane-embedded proton-conducting unit F0, forms the H+-ATP synthase that reversibly couples transmembrane proton flow to ATP synthesis/hydrolysis in respiring and photosynthetic cells. It has been suggested that the gamma-subunit of F1-ATPase rotates within the alphabeta-hexamer, a conjecture supported by structural, biochemical and spectroscopic studies. We attached a fluorescent actin filament to the gamma-subunit as a marker, which enabled us to observe this motion directly. In the presence of ATP, the filament rotated for more than 100 revolutions in an anticlockwise direction when viewed from the 'membrane' side. The rotary torque produced reached more than 40 pN nm(-1) under high load.

Stepped versus continuous rotatory motors at the molecular scale

Nature invented molecular rotatory devices such as the flagellar motor and ATP synthase. Photoselection techniques have been frequently used to detect the rotational random walk of proteins but only rarely for the rotational drift of subunits in proteins. Pertinent theories predict an oscillatory behavior of the polarization anisotropy, r, for unidirectional rotational drift, as opposed to a monotonic relaxation of r for bidirectional random walk. The underlying assumption of an angular continuum is questionable for intersubunit rotation in proteins. We developed a theory for stepped rotatory devices. It predicts the damped oscillation of r under unidirectional drift. Damping increases with decreasing number of steps. For only three steps a quasi-monotonic relaxation of r is predicted for both random walk and drift. In photoselection experiments with active F-ATPase we observed the relaxation of r when a spectroscopic probe was attached to the central beta-subunit. This behavior is compatible with the expectation for a three-stepped rotatory device.

Solution structure of the N-terminal domain of the delta subunit of the E. coli ATPsynthase

NMR studies of the delta subunit of the Escherichia coli F1F0-ATPsynthase reveal that it consists of an N-terminal six alpha-helix bundle and a less well ordered C terminus. Both domains are part of one of two separate connections between F1 and F0.

Tn3 resolvase catalyses multiple recombination events without intermediate rejoining of DNA ends

Resolvases and DNA invertases catalyse site-specific recombination by a concerted cut-and-religate mechanism. Topological data strongly suggest a rotational movement of the DNA half-sites during recombination: in an "iterative" mode of reaction, after cleavage of all four strands of the two recombining sites, the recombinase-linked half-sites seem to rotate through multiple steps of 180 degrees prior to final religation. However, current structural data provide no clear support for the postulated corresponding rotation of enzyme subunits within an active tetramer. A key issue is whether repetition of apparent 180 degrees rotation steps requires rejoining of the DNA strands and resetting of the catalytic machinery, or if multiple rotation steps can take place in the fully cleaved intermediate. We present evidence that a resolvase-catalysed DNA knotting reaction, brought about by apparent 360 degrees rotation, can proceed without rejoining of the DNA strands in the recombinant (180 degrees rotation) configuration. This behaviour is not compatible with a mechanism requiring a fixed arrangement of the catalytic subunits, and strongly suggests that recombination is coupled to disruption of the dimer interface between two subunits bound at each crossover site. We also show that an artificial supercoiled plasmid containing two res sites, with a single mismatched base-pair in one of the crossover sites, is a substrate for "suicidal" reactions in which resolvase remains covalently linked to two half-sites.